[Prev][Next][Index][Thread]
Re: 300W 54" TC specs: My questions
Hi Reinhard, all,
> I have a few questions to the design of your 54 incher.
>
> Here goes:
>
> Secondary: thickwall PVC drainpipe 9.9" diameter
> 980 turns 0.56mm copper, spacewound 1:1
> Ls about 45mH
> Cs about 19pF
> 9" sphere mounted several inches above windings
> Ctot =3D 26.3pF
> Fr =3D 146.5kHz
> Q unloaded about 300
>
> a.) Why did you space wind this coil. With a 9.9" diameter (and the
> corresponding length) I wouldn=B4t think this would be necessary. Wouldn=
> =B4t the Q
> be higher (higher inductance) on a non spacewound coil?
No. You have to take into account the proximity effect of the
windings which causes current bunching into a smaller cross-
sectional area. You can think of a closewound coil as a spacewound
one using a larger diameter wire. Essentially you are throwing copper
away - that's how it seems anyway.
> b.) Is there any way one can forsee how high above the secondary you shoul=
> d
> mount the toroid (for best performance)?
There should be a rule but I haven't found it yet. It would require
modelling based on coil size and geometry and topload size and
geometry as well as the winding details. I do know that the terminal
does need to be raised some distance to hit maximum Q for a
particular terminal size. The height appears to scale with coil and
terminal size.
> you calculate the Q of a loaded/unloaded coil?
Sorry for the unintentional snip. I can't because it is affected by
so many things. However, there is a guide available based on some
comprehensive measurements I took on a number of coils with a high
degree of isolation. Dr Rzezsotarksi has the details handy I believe.
> My coil is very similar to yours (electrical specs, but different former, =
> wire
> size)
> --------------------------------------------------------------------------=
> ----
> ------------------
> My secondary coil specs:
> Secondary wire: 0.85 mm =3D AWG "19.5"
> Wire spool weight before/after winding: 22lbs/12lbs
> i.d:0.85mm=3D0.0334567"
> o.d.:0.908mm=3D0.0357395"
>
> Secondary former:
> Diameter:20 cm=3D7.8722"
> Total length:106 cm
> Chosen winding length:82.85 cm=3D32.61"
> h/d ratio: 4.12
> Turns:940+1.2 (space wound from end of winding to end of former)
> Aprox wire length: 590m=3D1937 ft.
> DC resistance=3D 17.0 ohms
> Calc=B4d inductance: 37.86 mH
> Measured 35.6 mH
> (pretty damn close I would say)
> Cself=3D14.68 pF
> Ctoroid should be somewhere between 14.68pF and 58.73 pF
> So, I chose a Al dryer duct toroid with:
> Tube diameter: 4.271"
> Outer toroid diameter:31.8"
> which gives me a Ctoroid of 30.78pF (about midway between)
> My loaded resonance frequency is 121 khz, which should be good for some lo=
> ng
> sparks. We will see, however. (Calcs where made with my homebrew tc progra=
> m)
> --------------------------------------------------------------------------=
> ----
> -----------------------------------
The current topload on my coil is far too small for optimum
performance.
> Now back to your coil:
>
> Primary: 3 turns 3/8" Cu tubing helical
> diameter =3D 26"
> height =3D 3"
> Lp about 11 - 12uH
>
> 1.) Doesn=B4t a helical coil give you a hell of a lot (read: too much) of
> coupling (for this relatively big coil)
Again it is a geopmetry related thing. Think of the proximity of the
two coils to one another coupled with the external secondary area
covered by the primary. The system measures k=0.12
> 2.) I didn=B4t want to go for a helical (on my smaller coils I have used t=
> hese
> with great sucess) coil on my 8", fearing I might get lots of arc overs fr=
> om
> the secondary to the primary. I wanted to go for either a conical (harder =
> to
> wind, but a nice "inbetweener" on coupling factor) or a flat primary.
> Comments?
Flahsovers are always a problem for short coils. Ultimately, output
voltage is the final arbiter.
> Cap: dry fired 0.1uF poly.
>
> 1.) Is this a (single) homemade cap? If so how did you get the nice high 1=
> 00nF
> value? What material (sizes, etc) did you use.
Details of construction are online on Jim Fosse's homepage. There are
(from memory) 12 plates, each 1 metre long. Dielectric between the
plates consists of 2 layers of 0.25mm polyethylene. A 12 metre long
roll of cooking foil just does the job.
> Vgap set to 7 - 8kV
> Gap: single static, 1/2" rounded tungsten carbide tipped brass.
>
> 1.) Did you ever try using multiple spark gaps? I have always seen improve=
> ment
> on using more than one spark gap. All my experiments with vacuum or pressu=
> re
> aided quenching on single (a la Gary Lau) or multiple spark gaps have alwa=
> ys
> ended with miserable results, so (up to now) I have stayed with the flat
> multiple gap design (no forced air flow). I planned on using 35 mm copper =
> pipe
> for my 8"er. I might go for a SRSG in the end, don=B4t know yet.
It runs very well with the current topload. Additional gaps add
additional conduction drops in the primary circuit - an important
consideration when running low primary voltages. I see no need to
change unless the topload does. The gap quenches first notch only on
attached discharges. The does not appear to detract from performance
at all.
> SNIP
> I made the mistake of running it about 10 feet away from the gap.
> This resulted in a winding flashover and me having to rewind about 7
> layers of the secondary that had burnt. It has never given a hint of
> trouble with this coil.
>
> Where was your safety gap during this run? At the transfomer or at the end=
> of
> the long wires (running to your tank setup)? If the safety gap was at the =
> far
> end of the wires (away from the xformer) I can understand why you got a fl=
> ash
> over.
No safety gap ever used. One has to be careful in making assumptions
about how and why such gaps fire. Remember, the only time the
transformer cooked was at the end of a long run from its secondary to
the gap. I used to think that the problem occurred when the gap first
fired, shock exciting the lines/chokes etc. This doesn't stand to
reason considering I use chokes in the coil mentioned here. I now
think it could be that residual energy left in the primary is
responsible when the gap opens. I still don't really know. I do know
that back coupling from an un-empty secondary with k set too high can
excite safety gaps and that loosening the coupling can make the
problem go away. It is definitely a problem begging for a quantified
answer.
Malcolm